Performing a 10-fold cross validation - Revision history

Keilwagen at 07:51, 6 June 2011

2011-06-06T07:51:55Z

← Older revision		Revision as of 07:51, 6 June 2011
Line 37:		Line 37:
	1,		1,
	//stopping criterion		//stopping criterion
	1E-6,		new SmallDifferenceOfFunctionEvaluationsCondition(1E-6),
	//parameterization of the model, LAMBDA complies with the		//parameterization of the model, LAMBDA complies with the
	//parameterization by log-probabilities		//parameterization by log-probabilities

Grau: Undo revision 386 by 59.165.249.102 (talk)

2011-05-11T16:32:08Z

Undo revision 386 by 59.165.249.102 (talk)

← Older revision		Revision as of 16:32, 11 May 2011
Line 1:		Line 1:
	~~Sonuds great~~ to ~~me BWTHDIK~~		<source lang="java5">
			//create a Sample for each class from the input data, using the DNA alphabet
			Sample[] data = new Sample[2];
			data[0] = new DNASample( args[0] );

			//the length of our input sequences
			int length = data[0].getElementLength();

			data[1] = new Sample( new DNASample( args[1] ), length );

			AlphabetContainer container = data[0].getAlphabetContainer();

			//create a new PWM
			BayesianNetworkModel pwm = new BayesianNetworkModel( new BayesianNetworkModelParameterSet(
			//the alphabet and the length of the model:
			container, length,
			//the equivalent sample size to compute hyper-parameters
			4,
			//some identifier for the model
			"my PWM",
			//we want a PWM, which is an inhomogeneous Markov model (IMM) of order 0
			ModelType.IMM, (byte) 0,
			//we want to estimate the MAP-parameters
			LearningType.ML_OR_MAP ) );

			//create a new mixture model using 2 PWMs
			MixtureModel mixPwms = new MixtureModel(
			//the length of the mixture model
			length,
			//the two components, which are PWMs
			new Model[]{pwm,pwm},
			//the number of starts of the EM
			10,
			//the equivalent sample sizes
			new double[]{pwm.getESS(),pwm.getESS()},
			//the hyper-parameters to draw the initial sequence-specific component weights (hidden variables)
			1,
			//stopping criterion
			1E-6,
			//parameterization of the model, LAMBDA complies with the
			//parameterization by log-probabilities
			Parameterization.LAMBDA);

			//create a new inhomogeneous Markov model of order 3
			BayesianNetworkModel mm = new BayesianNetworkModel(
			new BayesianNetworkModelParameterSet( container, length, 256, "my iMM(3)", ModelType.IMM, (byte) 3, LearningType.ML_OR_MAP ) );

			//create a new PWM scoring function
			BayesianNetworkScoringFunction dPwm = new BayesianNetworkScoringFunction(
			//the alphabet and the length of the scoring function
			container, length,
			//the equivalent sample size for the plug-in parameters
			4,
			//we use plug-in parameters
			true,
			//a PWM is an inhomogeneous Markov model of order 0
			new InhomogeneousMarkov(0));

			//create a new mixture scoring function
			MixtureScoringFunction dMixPwms = new MixtureScoringFunction(
			//the number of starts
			2,
			//we use plug-in parameters
			true,
			//the two components, which are PWMs
			dPwm,dPwm);

			//create a new scoring function that is an inhomogeneous Markov model of order 3
			BayesianNetworkScoringFunction dMm = new BayesianNetworkScoringFunction(container, length, 4, true, new InhomogeneousMarkov(3));

			//create the classifiers
			int threads = AbstractMultiThreadedOptimizableFunction.getNumberOfAvailableProcessors();
			AbstractScoreBasedClassifier[] classifiers = new AbstractScoreBasedClassifier[]{
			//model based with mixture model and Markov model
			new ModelBasedClassifier( mixPwms, mm ),
			//conditional likelihood based classifier
			new MSPClassifier( new GenDisMixClassifierParameterSet(container, length,
			//method for optimizing the conditional likelihood and
			//other parameters of the numerical optimization
			Optimizer.QUASI_NEWTON_BFGS, 1E-2, 1E-2, 1, true, KindOfParameter.PLUGIN, false, threads),
			//mixture scoring function and Markov model scoring function
			dMixPwms,dMm )
			};

			//create an new k-fold cross validation using above classifiers
			KFoldCrossValidation cv = new KFoldCrossValidation( classifiers );

			//we use a specificity of 0.999 to compute the sensitivity and a sensitivity of 0.95 to compute FPR and PPV
			MeasureParameters mp = new MeasureParameters(false, 0.999, 0.95, 0.95);
			//we do a 10-fold cross validation and partition the data by means of the number of symbols
			KFoldCVAssessParameterSet cvpars = new KFoldCVAssessParameterSet(PartitionMethod.PARTITION_BY_NUMBER_OF_SYMBOLS, length, true, 10);

			//compute the result of the cross validation and print them to System.out
			System.out.println( cv.assess( mp, cvpars, data ) );
			</source>

Grau: Undo revision 393 by 175.41.247.122 (talk)

2011-05-11T16:31:48Z

Undo revision 393 by 175.41.247.122 (talk)

Can't load revision 442

at 04:48, 11 May 2011

2011-05-11T04:48:21Z

QCdzPifOfqgm

Can't load revision 393

at 02:26, 11 May 2011

2011-05-11T02:26:34Z

sBWqcJPHEVXqbGPW

Can't load revision 386

Keilwagen at 14:27, 2 December 2009

2009-12-02T14:27:21Z

← Older revision		Revision as of 14:27, 2 December 2009
Line 1:		Line 1:
	<source lang="java5">		<source lang="java5">
	//create a DNA-alphabet		//create a Sample for each class from the input data, using the DNA alphabet
	~~AlphabetContainer container~~ = new ~~AlphabetContainer(~~ new ~~DNAAlphabet~~() );		Sample[] data = new Sample[2];
			data[0] = new DNASample( args[0] );

	//the length of our input sequences		//the length of our input sequences
	int length = 7;		int length = data[0].getElementLength();

	~~//create a Sample for each class from the input~~ data~~, using the alphabet from above~~		data[1] = new Sample( new DNASample( args[1] ), length );
	~~Sample~~[] ~~data~~ = new Sample~~[]{~~
	~~new Sample( container, new StringExtractor~~( new ~~File~~(args[0]), ~~100~~ ) ),		AlphabetContainer container = data[0].getAlphabetContainer();
	~~new Sample(~~ container~~, new StringExtractor( new File(args~~[1])~~, 100 ), length ) }~~;

	//create a new PWM		//create a new PWM
Line 22:		Line 23:
	//we want to estimate the MAP-parameters		//we want to estimate the MAP-parameters
	LearningType.ML_OR_MAP ) );		LearningType.ML_OR_MAP ) );

	//create a new mixture model using 2 PWMs		//create a new mixture model using 2 PWMs
	MixtureModel mixPwms = new MixtureModel(		MixtureModel mixPwms = new MixtureModel(
Line 40:		Line 41:
	//parameterization by log-probabilities		//parameterization by log-probabilities
	Parameterization.LAMBDA);		Parameterization.LAMBDA);

	//create a new inhomogeneous Markov model of order 3		//create a new inhomogeneous Markov model of order 3
	BayesianNetworkModel mm = new BayesianNetworkModel(		BayesianNetworkModel mm = new BayesianNetworkModel(
	new BayesianNetworkModelParameterSet( container, length, 256, "my iMM(3)", ModelType.IMM, (byte) 3, LearningType.ML_OR_MAP ) );		new BayesianNetworkModelParameterSet( container, length, 256, "my iMM(3)", ModelType.IMM, (byte) 3, LearningType.ML_OR_MAP ) );

	//create a new PWM scoring function		//create a new PWM scoring function
	BayesianNetworkScoringFunction dPwm = new BayesianNetworkScoringFunction(		BayesianNetworkScoringFunction dPwm = new BayesianNetworkScoringFunction(
Line 55:		Line 56:
	//a PWM is an inhomogeneous Markov model of order 0		//a PWM is an inhomogeneous Markov model of order 0
	new InhomogeneousMarkov(0));		new InhomogeneousMarkov(0));

	//create a new mixture scoring function		//create a new mixture scoring function
	MixtureScoringFunction dMixPwms = new MixtureScoringFunction(		MixtureScoringFunction dMixPwms = new MixtureScoringFunction(
	//the number of starts		//the number of starts
	10,		2,
	//we use plug-in parameters		//we use plug-in parameters
	true,		true,
	//the two components, which are PWMs		//the two components, which are PWMs
	dPwm,dPwm);		dPwm,dPwm);

	//create a new scoring function that is an inhomogeneous Markov model of order 3		//create a new scoring function that is an inhomogeneous Markov model of order 3
	BayesianNetworkScoringFunction dMm = new BayesianNetworkScoringFunction(container, length, 4, true, new InhomogeneousMarkov(3));		BayesianNetworkScoringFunction dMm = new BayesianNetworkScoringFunction(container, length, 4, true, new InhomogeneousMarkov(3));

	//create the classifiers		//create the classifiers
			int threads = AbstractMultiThreadedOptimizableFunction.getNumberOfAvailableProcessors();
	AbstractScoreBasedClassifier[] classifiers = new AbstractScoreBasedClassifier[]{		AbstractScoreBasedClassifier[] classifiers = new AbstractScoreBasedClassifier[]{
	//model based with mixture model and Markov model		//model based with mixture model and Markov model
	new ModelBasedClassifier( mixPwms, mm ),		new ModelBasedClassifier( mixPwms, mm ),
	//conditional likelihood based classifier		//conditional likelihood based classifier
	new ~~CLLClassifier~~( new ~~CLLClassifierParameterSet~~(container, length,		new MSPClassifier( new GenDisMixClassifierParameterSet(container, length,
	//method for optimizing the conditional likelihood and		//method for optimizing the conditional likelihood and
	//other parameters of the numerical optimization		//other parameters of the numerical optimization
	Optimizer.QUASI_NEWTON_BFGS, 1E-6, 1E-6, ~~1E-2~~, true, KindOfParameter.PLUGIN, false),		Optimizer.QUASI_NEWTON_BFGS, 1E-2, 1E-2, 1, true, KindOfParameter.PLUGIN, false, threads),
	//mixture scoring function and Markov model scoring function		//mixture scoring function and Markov model scoring function
	dMixPwms,dMm )		dMixPwms,dMm )
	};		};

	//create an new k-fold cross validation using above classifiers		//create an new k-fold cross validation using above classifiers
	KFoldCrossValidation cv = new KFoldCrossValidation( classifiers );		KFoldCrossValidation cv = new KFoldCrossValidation( classifiers );

	//we use a specificity of 0.999 to compute the sensitivity and a sensitivity of 0.95 to compute FPR and PPV		//we use a specificity of 0.999 to compute the sensitivity and a sensitivity of 0.95 to compute FPR and PPV
	MeasureParameters mp = new MeasureParameters(false, 0.999, 0.95, 0.95);		MeasureParameters mp = new MeasureParameters(false, 0.999, 0.95, 0.95);
	//we do a 10-fold cross validation and partition the data by means of the number of symbols		//we do a 10-fold cross validation and partition the data by means of the number of symbols
	KFoldCVAssessParameterSet cvpars = new KFoldCVAssessParameterSet(PartitionMethod.PARTITION_BY_NUMBER_OF_SYMBOLS, length, true, 10);		KFoldCVAssessParameterSet cvpars = new KFoldCVAssessParameterSet(PartitionMethod.PARTITION_BY_NUMBER_OF_SYMBOLS, length, true, 10);

	//compute the result of the cross validation and print them to System.out		//compute the result of the cross validation and print them to System.out
	System.out.println( cv.assess( mp, cvpars, data ) );		System.out.println( cv.assess( mp, cvpars, data ) );
	</source>		</source>

Grau at 09:33, 29 October 2008

2008-10-29T09:33:44Z

← Older revision		Revision as of 09:33, 29 October 2008
Line 76:		Line 76:
	//method for optimizing the conditional likelihood and		//method for optimizing the conditional likelihood and
	//other parameters of the numerical optimization		//other parameters of the numerical optimization
	Optimizer.QUASI_NEWTON_BFGS, 1E-6, 1E-6, 1E-2, true, ~~true~~, false),		Optimizer.QUASI_NEWTON_BFGS, 1E-6, 1E-6, 1E-2, true, KindOfParameter.PLUGIN, false),
	//mixture scoring function and Markov model scoring function		//mixture scoring function and Markov model scoring function
	dMixPwms,dMm )		dMixPwms,dMm )

Keilwagen at 18:51, 12 September 2008

2008-09-12T18:51:27Z

← Older revision		Revision as of 18:51, 12 September 2008
Line 6:		Line 6:

	//create a Sample for each class from the input data, using the alphabet from above		//create a Sample for each class from the input data, using the alphabet from above
	Sample[] data = new Sample[]{ new Sample( container, new StringExtractor( new File(args[0]), 100 ) ),		Sample[] data = new Sample[]{
	new Sample( container, new StringExtractor( new File(args[1]), 100 ), length ) };		new Sample( container, new StringExtractor( new File(args[0]), 100 ) ),
			new Sample( container, new StringExtractor( new File(args[1]), 100 ), length ) };

	//create a new PWM		//create a new PWM

Keilwagen at 18:50, 12 September 2008

2008-09-12T18:50:32Z

← Older revision		Revision as of 18:50, 12 September 2008
Line 42:		Line 42:
	//create a new inhomogeneous Markov model of order 3		//create a new inhomogeneous Markov model of order 3
	BayesianNetworkModel mm = new BayesianNetworkModel(		BayesianNetworkModel mm = new BayesianNetworkModel(
	new BayesianNetworkModelParameterSet( container, length, 256, "my ~~PWM~~", ModelType.IMM, (byte) 3, LearningType.ML_OR_MAP ) );		new BayesianNetworkModelParameterSet( container, length, 256, "my iMM(3)", ModelType.IMM, (byte) 3, LearningType.ML_OR_MAP ) );

	//create a new PWM scoring function		//create a new PWM scoring function

Grau at 12:21, 7 September 2008

2008-09-07T12:21:28Z

← Older revision		Revision as of 12:21, 7 September 2008
Line 1:		Line 1:
	<source lang="~~java~~">		<source lang="java5">
	//create a DNA-alphabet		//create a DNA-alphabet
	AlphabetContainer container = new AlphabetContainer( new DNAAlphabet() );		AlphabetContainer container = new AlphabetContainer( new DNAAlphabet() );